Tunnel construction model building method and system

Abstract

The invention relates to the technical field of tunnel construction, and discloses a tunnel construction model establishment method and system, and the method comprises the steps: dividing a tunnel construction model into a plurality of thin sub-models; a model thin layer range is determined according to the current construction section information, a corresponding current thin layer sub-model is extracted, a current thumbnail model is formed through thumbnail, and a real-time decision driving module used for carrying out construction decision making on the current thumbnail model is established; construction monitoring data generated in the construction process are rapidly injected into the real-time decision driving module; based on the injected construction monitoring data, carrying out multi-source data fusion and data transmission in a real-time decision driving module so as to carry out abnormal risk identification processing; and obtaining intelligent decision data, and pushing the intelligent decision data to each thin-layer sub-model in the current thumbnail model. The problem that a tunnel construction model cannot make intelligent decision support in real time according to actual construction monitoring data is solved.

Description

Technical Field

[0001] This invention relates to the field of tunnel construction technology, and in particular to a method for establishing a tunnel construction model and a system for establishing a tunnel construction model. Background Technology

[0002] In tunnel construction, establishing an accurate construction model is crucial for guiding construction and controlling risks. Current technologies require building a model encompassing the entire tunnel route, integrating design information, geological conditions, construction progress, and resource information. While this model offers comprehensive information, it suffers from significant drawbacks: the sheer number of model elements results in a massive computational burden, with a single complete numerical simulation potentially taking hours or even days; the long model update cycle fails to meet the demands of real-time on-site decision-making; and the disconnect between data acquisition and model update processes hinders the timely integration of real-time monitoring data into the model.

[0003] Therefore, based on existing model building methods, there is a lack of effective feedback mechanism between the large amount of real-time data generated during construction (such as TBM sensor data, monitoring data, and geological sketches) and the construction model. Data acquisition requires multiple stages, including transmission, processing, and analysis, before it can be fed back into the model. Due to the massive amount of data in a complete tunnel construction model, this process typically takes several hours or even days. When the TBM encounters adverse geological conditions during excavation, sensor data cannot be fed back into the model in a timely manner, and by then, the excavation may have already entered a high-risk area, missing the optimal opportunity for intervention.

[0004] In addition, data collection during tunnel construction requires manual data processing, format conversion, and model parameter adjustment, which is inefficient and prone to errors. Data collected from different sources (geological data, monitoring data, and construction data) uses different data standards and formats, making effective integration difficult and hindering data correlation and utilization.

[0005] Therefore, the drawbacks of existing tunnel construction model establishment methods are: due to the large amount of model data, the amount of computation is huge and the calculation cycle is long, which makes it impossible for the tunnel construction model to make intelligent decision support based on actual construction monitoring data in real time. Summary of the Invention

[0006] The main objective of this invention is to provide a method and system for establishing a tunnel construction model, which aims to solve the problem in the prior art that the tunnel construction model cannot make intelligent decision support based on actual construction monitoring data in real time.

[0007] To achieve the above objectives, the present invention provides a method for establishing a tunnel construction model, comprising the following steps: Obtain the established tunnel construction model and divide it into multiple thin-layer sub-models with spatial connection order; Obtain information about the current construction section and determine the range of the thin layer in the model based on the information about the current construction section; Based on the location range of the thin layer in the tunnel construction model, the corresponding current thin layer sub-model is extracted, and the current abbreviated model is formed by abbreviating the model parameters of the current thin layer sub-model. A real-time decision-driven module for making construction decisions based on the current abbreviated model is also established. An automated data interface for acquiring construction monitoring data is established, which quickly injects the construction monitoring data generated during the construction process into the real-time decision-driven module. The construction monitoring data includes TBM sensor data, surrounding rock deformation monitoring data, support structure stress monitoring data, environmental monitoring data, geological sketch data, and advanced geological forecast data. Based on the injected construction monitoring data, multi-source data fusion and data transmission are performed in the real-time decision-driven module to identify and process abnormal risks. Based on the results of abnormal risk identification, intelligent decision-making data is obtained and pushed to each thin-layer sub-model in the current abbreviated model.

[0008] Optionally, the step of obtaining the established tunnel construction model and dividing the tunnel construction model into multiple thin-layer sub-models with spatial connection order includes: Obtain the geological survey data and construction process parameters preset in the tunnel construction model; Based on the gradient of geological conditions and the frequency of construction technology adjustments, the boundary is dynamically searched and delineated in the tunnel construction model. Based on each boundary, the tunnel construction model is divided into multiple thin-layer sub-models with spatial connection order.

[0009] Optionally, the step of dynamically searching and defining boundaries in the tunnel construction model based on the gradient of geological condition changes and the frequency of construction technology adjustments includes: The sliding window algorithm is adopted to slide the window within the tunnel construction model interval corresponding to the recent construction section, and calculate the gradient of geological condition changes and the frequency of construction technology adjustment within the window. The recent construction section includes the excavated but unstable section and the section to be constructed. When the gradient of geological conditions or the frequency of construction process adjustments within the window exceeds the threshold, the boundary of the thin-layer sub-model is determined according to the corresponding window position. Adjust the position, length, and shape of the boundary according to the preset adjustment range.

[0010] Optionally, the step of obtaining the current construction section information and determining the range of the thin layer of the model based on the current construction section information includes: Based on the current location of the construction section and the tunneling speed, predict the range of construction progress within the preset time frame in the future; Based on the predicted construction progress range and combined with the trend of geological condition changes, the front and rear boundaries of the model thin layer range are dynamically determined. Based on construction monitoring data, the risk-affected areas are detected, and when adverse geological conditions or construction anomalies are encountered, the model's thin-layer range is automatically expanded to include the risk-affected areas.

[0011] Optionally, the step of extracting the corresponding current thin-layer sub-model based on the location interval of the model thin-layer range within the tunnel construction model, and forming the current abbreviated model based on the model parameters of the current thin-layer sub-model, includes: Based on the location interval of the thin layer range in the tunnel construction model, extract several consecutive thin layer sub-models as the current thin layer sub-model; Based on the surrounding rock grade, deformation rate, groundwater conditions, and construction stage in the current thin-layer sub-model, the current thin-layer sub-model is reduced to form the current abbreviated model.

[0012] Optionally, the step of establishing a real-time decision-driven module for making construction decisions based on the current scaled-down model includes: Based on the thin-layer sub-models contained in the current abbreviated model, establish the driving space of the real-time decision-driven module; Based on the spatial topology of each thin-layer sub-model in the current abbreviated model, a data transmission mechanism for each thin-layer sub-model is established in the real-time decision-driven module to realize the real-time sharing of construction monitoring data and predicted construction monitoring data of each thin-layer sub-model in the real-time decision-driven module; wherein, the spatial topology includes the forward connection relationship, backward connection relationship and lateral adjacent connection relationship of each thin-layer sub-model in the current abbreviated model; Establish a shared data storage area corresponding to the driving space of the real-time decision-driven module.

[0013] Optionally, the step of performing multi-source data fusion and data transmission in the real-time decision-driven module based on the injected construction monitoring data for anomaly risk identification and processing includes: Temporal variation data is extracted from construction monitoring data, including TBM sensor temporal data, surrounding rock deformation temporal data, support structure stress temporal data, environmental monitoring temporal data, geological sketch temporal data, and advanced geological prediction temporal data. The real-time decision-driven module performs multi-source feature fusion on time-series changing data, injects the fused multi-source features into the data storage area corresponding to the thin-layer sub-model where the working face is located, and transmits data to other thin-layer sub-models in the current abbreviated model according to the data transmission mechanism, generating predictive construction monitoring data and injecting it into the corresponding data storage area, thereby generating decision-driven data packages. Anomaly risk identification and processing based on decision-driven data packets.

[0014] Optionally, the step of transferring data to other thin-layer sub-models in the current scaled-down model according to the data transfer mechanism, and generating predictive construction monitoring data for injection into the corresponding data storage area, includes: Based on the forward connectivity, backward connectivity, lateral adjacency connectivity and corresponding connectivity weights of each thin-layer sub-model in the current abbreviated model in the spatial topology, as well as the fused multi-source features, the predicted construction monitoring data corresponding to each of the remaining thin-layer sub-models in the current abbreviated model are calculated to achieve data transmission. The predicted construction monitoring data corresponding to the other thin-layer sub-models in the current abbreviated model are fused into predicted multi-source features and stored in the data storage area corresponding to the other thin-layer sub-models in the current abbreviated model to achieve data injection.

[0015] Optionally, the step of performing anomaly risk identification processing based on decision-driven data packets includes: The real-time decision-driven module inputs the decision-driven data package into the risk type prediction model and uses the risk type prediction model to predict the current construction status type. The construction status types include normal status, surrounding rock instability, water and mud inrush, collapse and roof fall, equipment failure and surface subsidence. Based on the predicted construction status type, identify whether there are any abnormal risks.

[0016] To achieve the above objectives, the present invention also proposes a tunnel construction model establishment system, wherein the tunnel construction model establishment system applies the tunnel construction model establishment method; the tunnel construction model establishment system includes: The abbreviated model module is used to acquire the established tunnel construction model, divide the tunnel construction model into multiple thin-layer sub-models with spatial connection order; acquire the current construction segment information, determine the thin-layer range of the model based on the current construction segment information; extract the corresponding current thin-layer sub-model based on the location interval of the thin-layer range of the model in the tunnel construction model, abbreviate the current abbreviated model based on the model parameters of the current thin-layer sub-model, and establish a real-time decision-driven module for making construction decisions based on the current abbreviated model; The identification module is used to establish an automated data interface for acquiring construction monitoring data, and to quickly inject the construction monitoring data generated during the construction process into the real-time decision-driven module. The construction monitoring data includes TBM sensor data, surrounding rock deformation monitoring data, support structure stress monitoring data, environmental monitoring data, geological sketch data, and advanced geological forecast data. Based on the injected construction monitoring data, the real-time decision-driven module performs multi-source data fusion and data transmission to identify and process abnormal risks. The decision-making module is used to obtain intelligent decision-making data based on the results of abnormal risk identification, and to push the intelligent decision-making data to each thin-layer sub-model in the current abbreviated model.

[0017] The technical solution of this invention helps to solve the problem in existing technologies where tunnel construction models cannot provide intelligent decision support in real time based on actual construction monitoring data. Specifically, the invention divides the established tunnel construction model into multiple thin-layer sub-models, distributing the massive data volume of the original tunnel construction model across each thin-layer sub-model. Based on the current tunnel construction progress, the corresponding thin-layer range is determined. The thin-layer sub-model corresponding to the thin-layer range, along with the corresponding model parameters, is used to abbreviate the model, further reducing the data volume to form the current abbreviated model. This abbreviated model uses a corresponding real-time decision-driven module for decision-making. Construction monitoring data generated during construction is rapidly injected into the real-time decision-driven module. The real-time decision-driven module performs multi-source data fusion and data transmission on the injected construction monitoring data to identify abnormal risks. Therefore, when an abnormal risk is identified, intelligent decision data can be quickly obtained and pushed to each thin-layer sub-model in the current abbreviated model. Thus, this invention facilitates intelligent decision support in real time based on actual construction monitoring data. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the tunnel construction model establishment method in the first embodiment of the present invention; Figure 2 This is a schematic diagram of the thin-layer sub-model in this invention; Figure 3 This is a schematic diagram of the thin layer range of the model corresponding to the current construction section information in this invention; Figure 4 This is a schematic diagram of the spatial topology of the thin-layer sub-model where the working face is located in the current abbreviated model of this invention.

[0019] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0021] In the following description, the use of suffixes such as "unit," "component," or "element" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "unit," "component," or "element" may be used interchangeably.

[0022] Please see Figures 1 to 4The first embodiment of the present invention provides a method for establishing a tunnel construction model, comprising the following steps: Step S10: Obtain the established tunnel construction model and divide the tunnel construction model into multiple thin-layer sub-models with spatial connection order. Step S20: Obtain the current construction section information and determine the thin layer range of the model based on the current construction section information; Step S30: Based on the location range of the thin layer in the tunnel construction model, extract the corresponding current thin layer sub-model, reduce the current abbreviated model based on the model parameters of the current thin layer sub-model, and establish a real-time decision-driven module for making construction decisions based on the current abbreviated model. Step S40: Establish an automated data interface for acquiring construction monitoring data, and quickly inject the construction monitoring data generated during the construction process into the real-time decision-driven module; wherein, the construction monitoring data includes TBM sensor data, surrounding rock deformation monitoring data, support structure stress monitoring data, environmental monitoring data, geological sketch data and advanced geological prediction data; Step S50: Based on the injected construction monitoring data, multi-source data fusion and data transmission are performed in the real-time decision-driven module to identify and process abnormal risks. Step S60: Based on the abnormal risk identification results, obtain intelligent decision data and push the intelligent decision data to each thin-layer sub-model in the current abbreviated model.

[0023] The technical solution of this invention helps to solve the problem in existing technologies where tunnel construction models cannot provide intelligent decision support in real time based on actual construction monitoring data. Specifically, the invention divides the established tunnel construction model into multiple thin-layer sub-models, distributing the massive data volume of the original tunnel construction model across each thin-layer sub-model. Based on the current tunnel construction progress, the corresponding thin-layer range is determined. The thin-layer sub-model corresponding to the thin-layer range, along with the corresponding model parameters, is used to abbreviate the model, further reducing the data volume to form the current abbreviated model. This abbreviated model uses a corresponding real-time decision-driven module for decision-making. Construction monitoring data generated during construction is rapidly injected into the real-time decision-driven module. The real-time decision-driven module performs multi-source data fusion and data transmission on the injected construction monitoring data to identify abnormal risks. Therefore, when an abnormal risk is identified, intelligent decision data can be quickly obtained and pushed to each thin-layer sub-model in the current abbreviated model. Thus, this invention facilitates intelligent decision support in real time based on actual construction monitoring data.

[0024] Specifically, tunnel construction models can simulate the entire process of tunnel excavation, support structure construction, and surrounding rock deformation, predicting the stress distribution, deformation characteristics, and stability state of the surrounding rock at different construction stages, and providing a basis for optimizing construction plans.

[0025] TBM sensor data refers to the sensor data of the tunnel boring machine.

[0026] Construction monitoring data is used for feedback of construction operation parameters at the construction site, judgment of surrounding rock stability (including surrounding rock deformation, stress changes, etc.), changes in the working status of support structures, feedback of environmental conditions, and feedback of geological data.

[0027] Among them, TBM sensor data, surrounding rock deformation monitoring data, support structure stress monitoring data and environmental monitoring data can be directly collected through the corresponding sensors, while geological sketch data and advanced geological forecast data can be entered using mobile terminals based on on-site observation and judgment by construction personnel.

[0028] The current abbreviated model includes several thin-layer sub-models of the tunnel construction model, which significantly reduces the inconvenience of loading the complete tunnel construction model on the construction site. It also further abbreviates the data of the thin-layer sub-models, compressing the complete tunnel model from millions of components to thousands to tens of thousands of components based on the current construction section, reducing the data volume by up to 90%. Employing a memory-mapping-based fast data loading technology, the model loading time is reduced from minutes to seconds, supporting smooth viewing on mobile devices.

[0029] Furthermore, to enable localized injection of the current scaled-down model, edge computing nodes can be deployed at the tunnel entrance, injecting only the construction monitoring data of the current construction segment into the current scaled-down model. A breakpoint resume mechanism is employed, with data cached locally during network interruptions and automatically synchronized upon network recovery to ensure data integrity.

[0030] The current thin-layer sub-model includes at least one thin-layer sub-model.

[0031] According to the first embodiment of the tunnel construction model establishment method of the present invention, and the second embodiment of the tunnel construction model establishment method of the present invention, step S10 includes: Step S11: Obtain the preset geological survey data and construction process parameters in the tunnel construction model; Step S12: Adjust the frequency according to the gradient of geological conditions and construction technology, and dynamically search and delineate the boundaries in the tunnel construction model; Step S13: Based on each boundary, divide the tunnel construction model into multiple thin-layer sub-models with spatial connection order.

[0032] Specifically, geological survey data includes the physical and mechanical parameters of the surrounding rock, geological structural characteristics, geostress field parameters, and information on adverse geological bodies; construction process parameters include tunneling parameters, support parameters, and grouting parameters.

[0033] Dynamically adjusting the boundary of the thin-layer sub-model based on the gradient of geological conditions and the frequency of construction technology adjustment is beneficial for classifying similar tunnel areas into the same thin-layer sub-model.

[0034] The physical and mechanical properties, geological structural features, stress field, and information on adverse geological bodies of the surrounding rock exhibit significant spatial heterogeneity. This invention employs small-sized elements to capture local deformation characteristics in areas with drastic changes in surrounding rock conditions (such as fault fracture zones, weak interlayers, and lithological interfaces); while in areas with gradual changes in surrounding rock conditions, large-sized elements are used to reduce computational load. Adjusting the boundary division of the thin-layer sub-model according to the gradient of geological condition changes can better adapt to the heterogeneity of surrounding rock conditions.

[0035] Construction process parameters need to be dynamically adjusted based on measured surrounding rock conditions. In areas where surrounding rock conditions change drastically, the frequency of construction process adjustments is higher, requiring a more refined model to support construction decisions; while in areas where surrounding rock conditions change gradually, the construction process is relatively stable, and a coarser model can be used. Adjusting the boundary delineation of the thin-layer sub-model according to the frequency of construction process adjustments can match the model accuracy with construction requirements.

[0036] In this invention, the surrounding rock conditions and construction technology are updated as tunneling progresses. The boundary of the thin-layer sub-model is dynamically adjusted according to the gradient of geological condition changes and the frequency of construction technology adjustment. This enables the model to be adaptively updated, allowing the model to be adjusted to match the current construction state more accurately. The updated abbreviated model corresponding to the current construction progress is then injected into the construction monitoring data for real-time decision-making.

[0037] In a second embodiment of the tunnel construction model establishment method of the present invention, and in a third embodiment of the tunnel construction model establishment method of the present invention, step S12 includes: Step S121: Using the sliding window algorithm, the window is slid within the tunnel construction model interval corresponding to the recent construction section to calculate the gradient of geological condition changes and the frequency of construction technology adjustment within the window. The recent construction section includes the excavated but unstable section and the section to be affected by construction. Step S122: When the gradient of geological condition changes or the frequency of construction process adjustment within the window exceeds the threshold, the boundary of the thin-layer sub-model is determined according to the corresponding window position. Step S123: Adjust the position, length, and shape of the boundary according to the preset adjustment range.

[0038] Specifically, based on the geological survey data and construction process parameters preset in the tunnel construction model, the initial boundary division results of the thin-layer sub-model will be obtained, thereby obtaining the boundary position coordinates determined by the initial boundary division; For tunnel sections that have been excavated and supported (e.g., those that have been under stable construction for more than a preset time, such as 7 days), the corresponding thin-layer sub-model boundary will be locked, and the locked boundary will no longer participate in dynamic adjustments; for long-term unconstructed sections that are too far from the current tunnel face (e.g., more than 150 meters), the boundary will retain the result of the last update. Only recently constructed sections will have their boundaries dynamically updated.

[0039] The recent construction section includes the excavated but unstable section and the section to be affected by construction: the excavated but unstable section is the construction section within the first distance range in front of the current tunnel face and the second distance range behind it; the section to be affected by construction is the construction section from the third distance range in front of the current tunnel face to the first distance range in front, where the third distance is greater than the second distance. The first and second distance ranges are dynamically adjusted according to the actual project conditions (such as TBM tunneling speed and surrounding rock conditions), but usually do not exceed 150 meters.

[0040] When sliding the window within the tunnel construction model interval corresponding to the recent construction section, the window slides along the tunnel excavation direction. For the excavated but unstable section, reverse sliding is added to capture the surrounding rock deformation feedback effect.

[0041] When the gradient of geological conditions within the window is greater than the preset gradient value, or the frequency of construction process adjustment is greater than the preset frequency, the center position of the window will be determined as the boundary of the adjusted adjacent thin-layer sub-model.

[0042] Among them, the gradient of geological condition changes and the frequency of construction technology adjustments in the excavated but unstable section can be obtained through TBM sensor data, surrounding rock deformation monitoring data, support structure stress monitoring data, and geological sketch data; the gradient of geological condition changes in the section to be affected by construction can be obtained through advanced geological forecast data, and the frequency of construction technology adjustments in the section to be affected by construction is based on the results of advanced geological forecasts, predicting the engineering characteristics of the surrounding rock ahead, and predicting the adjustment needs of construction technology according to the surrounding rock grade and engineering experience.

[0043] In the first embodiment of the tunnel construction model establishment method of the present invention, and in the fourth embodiment of the tunnel construction model establishment method of the present invention, step S20 includes: Step S21: Based on the current construction section location and tunneling speed, predict the construction progress range within the future preset time. Step S22: Based on the predicted construction progress range and combined with the trend of geological condition changes, dynamically determine the front and rear boundaries of the thin layer range of the model. Step S23: Detect the risk-affected area based on construction monitoring data, so as to automatically expand the model thin layer range to include the risk-affected area when encountering adverse geological bodies or construction anomalies.

[0044] In step S21, based on the tunnel axis mileage coordinates, the coordinate range of each thin-layer sub-model and the spatial index relationship of each thin-layer sub-model are established. Thus, based on the current construction section location and tunneling speed, the construction progress range within the preset time period can be determined. By performing a spatial index query, the thin-layer sub-model to be extracted can be quickly located to determine the initial model thin-layer range.

[0045] Specifically, in step S22, dynamically determining the front and rear boundaries of the model's thin layer range based on the changing trends of geological conditions refers to determining the excavated unstable section and the section to be affected by construction within a preset range before and after the tunnel face, based on the tunnel face location. The lengths of both the section to be affected by construction and the excavated unstable section are determined based on changes in geological conditions and the frequency of construction technology adjustments. Smaller changes in geological conditions and lower frequency of construction technology adjustments result in shorter sections to be affected by construction; conversely, larger changes in geological conditions and higher frequency of construction technology adjustments result in longer sections to be affected by construction. Based on the determined range of the excavated unstable section and the section to be affected by construction, several consecutive thin layer sub-models within this range are extracted as the current thin layer sub-model, thus adjusting the front and rear boundaries of the model's thin layer range. Since changes in geological conditions and the frequency of construction technology adjustments may dynamically change with construction progress, the front and rear boundaries of the model's thin layer range will also be dynamically and adaptively adjusted accordingly.

[0046] Furthermore, in step S23, this embodiment also detects the risk-affected area based on construction monitoring data, so as to automatically expand the model thin layer range to include the risk-affected area when encountering adverse geological bodies or construction anomalies; equivalent to the fourth embodiment being able to adjust the model thin layer range to the maximum range affecting the current construction safety, which is beneficial to add all necessary thin layer sub-models to the current abbreviated model, so as to achieve rapid construction decision-making by injecting real-time construction monitoring data into the current abbreviated model.

[0047] In the first embodiment of the tunnel construction model establishment method of the present invention, and in the fifth embodiment of the tunnel construction model establishment method of the present invention, step S30, which involves extracting the corresponding current thin-layer sub-model based on the location interval of the model thin-layer range corresponding to the tunnel construction model, and forming a current abbreviated model based on the model parameters of the current thin-layer sub-model, includes: Step S31: Based on the location interval of the thin layer range in the tunnel construction model, extract several consecutive thin layer sub-models as the current thin layer sub-model. Step S32: Based on the surrounding rock grade, deformation rate, groundwater conditions and construction stage in the current thin-layer sub-model, the current thin-layer sub-model is reduced to form the current reduced model.

[0048] In this embodiment, the surrounding rock grade, deformation rate, groundwater conditions, and construction stage in the model parameters corresponding to the current thin-layer sub-model are used as indicators for the reduction of data volume within the sub-model.

[0049] Among them, the surrounding rock grade is classified according to the stability of the surrounding rock, and each surrounding rock grade corresponds to a first abbreviation factor with different value ranges; Deformation rate is a key indicator reflecting the stability of the current construction section. A second abbreviation factor with different value ranges is determined according to different deformation rates. Groundwater conditions directly affect the physical and mechanical properties of the surrounding rock and construction safety. A third abbreviation factor with different value ranges is determined based on different deformation rates. The fourth abbreviation factor, with different value ranges, is determined according to different construction stages.

[0050] Based on the first abbreviation factor, the second abbreviation factor, the third abbreviation factor, and the fourth abbreviation factor, the model abbreviation strategy for the current thin-layer sub-model is determined comprehensively.

[0051] According to the fifth embodiment of the tunnel construction model establishment method of the present invention, and the sixth embodiment of the tunnel construction model establishment method of the present invention, the establishment of a real-time decision-driving module for making construction decisions on the current abbreviated model in step S30 includes: Step S33: Based on the thin-layer sub-models contained in the current abbreviated model, establish the driving space of the real-time decision-driven module. Step S34: Based on the spatial topology of each thin-layer sub-model in the current abbreviated model, establish a data transmission mechanism between the real-time decision-driven module and each thin-layer sub-model, so as to realize the real-time sharing of construction monitoring data and predicted construction monitoring data of each thin-layer sub-model in the real-time decision-driven module; wherein, the spatial topology includes the forward connection relationship, backward connection relationship and lateral adjacent connection relationship of each thin-layer sub-model in the current abbreviated model; Step S35: Establish a shared data storage area corresponding to the driving space of the real-time decision-driven module.

[0052] Based on the forward, backward, and lateral adjacency relationships of each thin-layer sub-model in the current abbreviated model, the influence range of the construction monitoring data of the thin-layer sub-model where the tunnel face is located in the current abbreviated model can be determined in terms of its forward, backward, and adjacent thin-layer sub-models. This establishes the data transmission mechanism for the real-time decision-driven module. Furthermore, during construction, the thin-layer sub-models included in the current abbreviated model can be automatically updated according to the spatial topology. This is equivalent to adaptively updating the abbreviated range of the tunnel construction model and the driving space of the real-time decision-driven module. Thus, it can dynamically load new current abbreviated models as construction progresses, maintaining synchronization between the current abbreviated model and the construction progress, while also updating the driving space of the real-time decision-driven module in real time.

[0053] Specifically, when establishing forward connection relationships, the forward connection direction is determined based on the tunnel axis direction; a data transmission channel is established between the current construction section and the unconstructed section ahead; forward connection weights are set, and the influence of the thin-layer sub-model where the working face is located on the surrounding rock ahead is reflected by the forward connection weights, thereby enabling the forward transmission of construction monitoring data of the thin-layer sub-model where the current working face is located.

[0054] The backward connection relationship is used to establish a connection between the thin-layer sub-model where the tunnel face is located and the thin-layer sub-model in the direction of the constructed section. The establishment method is as follows: based on the tunnel axis direction, the backward connection direction is determined, a data transmission channel is established between the current construction section and the constructed section behind, and the backward connection weight is set. The backward connection weight reflects the degree of influence of the thin-layer sub-model where the tunnel face is located on the surrounding rock behind, thereby realizing the backward transmission of construction monitoring data of the thin-layer sub-model where the current tunnel face is located.

[0055] Lateral adjacency connection is the connection established between the thin-layer sub-model where the tunnel face is located and its laterally adjacent thin-layer sub-models. The establishment method includes: establishing the connection between the thin-layer sub-model where the tunnel face is located and its laterally adjacent thin-layer sub-models in the tunnel axis direction based on spatial adjacency; setting adjacency connection weights to reflect the interaction between the thin-layer sub-model where the tunnel face is located and its laterally adjacent thin-layer sub-models, thereby enabling the lateral transmission of construction monitoring data of the thin-layer sub-model where the current tunnel face is located.

[0056] In the data transmission mechanism of the real-time decision-driven module for each thin-layer sub-model, the transmitted content includes construction monitoring data, abnormal risk identification results, and rapid decision-making. Among them, the transmitted construction monitoring data includes TBM sensor data (TBM thrust, torque, rotation speed), surrounding rock deformation monitoring data (convergence deformation, settlement, displacement); support structure stress monitoring data (anchor bolt axial force, steel arch stress, shotcrete stress), environmental monitoring data (temperature, humidity, gas concentration), geological sketch data (lithology, joint density, faults, and groundwater), and advanced geological prediction data (geophysical data, drilling data, and geological inference data).

[0057] Data can be transmitted synchronously, asynchronously, or selectively (critical data can be selectively transmitted based on its importance).

[0058] The shared data storage area serves as the shared data storage for each thin-layer sub-model in the current abbreviated model; it provides a standardized data access interface and supports multi-threaded concurrent access.

[0059] After the real-time decision-driven module is established according to the above steps, it determines the size of the thin-layer subspace range (i.e., the number of thin-layer subspaces) where strategy decisions need to be made. The spatial topology then determines which thin-layer sub-models the real-time decision-driven module will apply the decision data to.

[0060] According to the sixth embodiment of the tunnel construction model establishment method of the present invention, and the seventh embodiment of the tunnel construction model establishment method of the present invention, step S50 includes: Step S51: Extract time-series variation data from construction monitoring data. The time-series variation data includes TBM sensor time-series data, surrounding rock deformation time-series data, support structure stress time-series data, environmental monitoring time-series data, geological sketch time-series data, and advanced geological prediction time-series data. Step S52: The real-time decision-driven module performs multi-source feature fusion on the time-series change data, injects the fused multi-source features into the data storage area corresponding to the thin-layer sub-model where the working face is located, and transmits data to other thin-layer sub-models in the current abbreviated model according to the data transmission mechanism, generating predictive construction monitoring data and injecting it into the corresponding data storage area, thereby generating a decision-driven data package. Step S53: Perform anomaly risk identification processing based on decision-driven data packets.

[0061] The time-series data from TBM sensors include the mean, variance, and trend of thrust, torque, and rotational speed; the time-series characteristics of surrounding rock deformation include convergence rate, settlement rate, and stress change rate; the time-series data of support structure stress include the rate of change of anchor bolt axial force, steel arch stress, and shotcrete stress; the time-series data of environmental monitoring include the rate of change of temperature, humidity, and gas concentration; the time-series characteristics of geological sketches include the rate of change of lithology, joint density, fault, and groundwater; and the time-series data of advanced geological prediction includes the rate of change of geophysical data, drilling data, and geological inference data. Specifically, raw construction monitoring data from different sources are spliced ​​together to form a high-dimensional feature vector.

[0062] The real-time decision-driven module stores high-dimensional feature vectors in a topological structure, specifically using an adjacency list or adjacency matrix storage space. This storage is used to record the connection relationships, connection weights, and data transmission directions of each thin-layer sub-model, and supports dynamic updates, automatically adjusting the topology structure as construction progresses.

[0063] Specifically, rapidly injecting construction monitoring data generated during construction into the real-time decision-driven module refers to injecting the construction monitoring data of the thin-layer sub-model at the current working face into the data storage area corresponding to the thin-layer sub-model at the current working face; while the other thin-layer sub-models in the current abbreviated model calculate and predict construction monitoring data according to the forward connection relationship, backward connection relationship, lateral adjacent connection relationship and corresponding connection weight, and realize data transmission according to the preset calculation logic, and inject the predicted construction monitoring data into the data storage area corresponding to each thin-layer sub-model to achieve shared data storage; Decision-driven data packages are generated based on the data injected into each data storage area.

[0064] When injecting data into the data storage area, a fast data loading technology based on memory mapping is used to load each piece of data into the shared data storage area.

[0065] In the seventh embodiment of the tunnel construction model establishment method of the present invention, and in the eighth embodiment of the tunnel construction model establishment method of the present invention, step S52, which involves transmitting data to the remaining thin-layer sub-models in the current scaled-down model according to the data transmission mechanism and generating predicted construction monitoring data to be injected into the corresponding data storage area, includes: Step S521: Based on the forward connectivity, backward connectivity, lateral adjacency connectivity and corresponding connection weights of each thin-layer sub-model in the current abbreviated model in the spatial topology, as well as the fused multi-source features, calculate the predicted construction monitoring data corresponding to each of the remaining thin-layer sub-models in the current abbreviated model, so as to realize data transmission. Step S522: The predicted construction monitoring data corresponding to the other thin-layer sub-models in the current abbreviated model are fused into predicted multi-source features and stored in the data storage area corresponding to the other thin-layer sub-models in the current abbreviated model to realize data injection.

[0066] During data transmission, the predicted construction monitoring data in the other thin-layer sub-models are calculated based on the connection weights, the fused multi-source features, and the preset calculation logic.

[0067] According to the seventh embodiment of the tunnel construction model establishment method of the present invention, and the ninth embodiment of the tunnel construction model establishment method of the present invention, step S53 includes: Step S531: The real-time decision-driven module inputs the decision-driven data package as input data into the risk type prediction model and uses the risk type prediction model to predict the current construction status type. The construction status types include normal status, surrounding rock instability, water and mud inrush, collapse and roof fall, equipment failure and surface subsidence. Step S532: Identify any abnormal risks based on the predicted construction status type.

[0068] Among them, the risk type prediction model determines the current construction status type based on the construction monitoring data of each thin-layer sub-model in the current abbreviated model (the model has a pre-stored mapping relationship between construction monitoring data and the current construction status type).

[0069] In the tenth embodiment of the tunnel construction model establishment method of the present invention, based on the first to ninth embodiments of the present invention, step S60 includes: Step S61: Based on the identified abnormal risks, obtain intelligent decision data that is mapped to each thin-layer sub-model of the current abbreviated model. Step S62: The intelligent decision data mapped to each thin-layer sub-model of the current abbreviated model is pushed to each thin-layer sub-model.

[0070] If the current risk can be corrected, the intelligent decision-making data can be used to optimize construction parameters such as tunneling speed, support parameters, and grouting pressure to improve construction efficiency while ensuring safety. If the current risk is difficult to correct, the intelligent decision-making data can be used to exit the current construction process.

[0071] To achieve the above objectives, the present invention also proposes a tunnel construction model establishment system, wherein the tunnel construction model establishment system applies the tunnel construction model establishment method; the tunnel construction model establishment system includes: The abbreviated model module is used to acquire the established tunnel construction model, divide the tunnel construction model into multiple thin-layer sub-models with spatial connection order; acquire the current construction segment information, determine the thin-layer range of the model based on the current construction segment information; extract the corresponding current thin-layer sub-model based on the location interval of the thin-layer range of the model in the tunnel construction model, abbreviate the current abbreviated model based on the model parameters of the current thin-layer sub-model, and establish a real-time decision-driven module for making construction decisions based on the current abbreviated model; The identification module is used to establish an automated data interface for acquiring construction monitoring data, and to quickly inject the construction monitoring data generated during the construction process into the real-time decision-driven module. The construction monitoring data includes TBM sensor data, surrounding rock deformation monitoring data, support structure stress monitoring data, environmental monitoring data, geological sketch data, and advanced geological forecast data. Based on the injected construction monitoring data, the real-time decision-driven module performs multi-source data fusion and data transmission to identify and process abnormal risks. The decision-making module is used to obtain intelligent decision-making data based on the results of abnormal risk identification, and to push the intelligent decision-making data to each thin-layer sub-model in the current abbreviated model.

[0072] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms, or by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to enter the methods described in the various embodiments of the present invention.

[0073] In the description of this specification, references to terms such as "one embodiment," "another embodiment," "other embodiments," or "first embodiment to Xth embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, method steps, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0074] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0075] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0076] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for establishing a tunnel construction model, characterized in that, Includes the following steps: Obtain the established tunnel construction model and divide it into multiple thin-layer sub-models with spatial connection order; Obtain information about the current construction section and determine the range of the thin layer in the model based on the information about the current construction section; Based on the location range of the thin layer in the tunnel construction model, the corresponding current thin layer sub-model is extracted, and the current abbreviated model is formed by abbreviating the model parameters of the current thin layer sub-model. A real-time decision-driven module for making construction decisions based on the current abbreviated model is also established. An automated data interface for acquiring construction monitoring data is established, which quickly injects the construction monitoring data generated during the construction process into the real-time decision-driven module. The construction monitoring data includes TBM sensor data, surrounding rock deformation monitoring data, support structure stress monitoring data, environmental monitoring data, geological sketch data, and advanced geological forecast data. Based on the injected construction monitoring data, multi-source data fusion and data transmission are performed in the real-time decision-driven module to identify and process abnormal risks. Based on the results of abnormal risk identification, intelligent decision-making data is obtained and pushed to each thin-layer sub-model in the current abbreviated model.

2. The method for establishing a tunnel construction model according to claim 1, characterized in that, The step of acquiring the established tunnel construction model and dividing the tunnel construction model into multiple thin-layer sub-models with spatial connection order includes: Obtain the geological survey data and construction process parameters preset in the tunnel construction model; Based on the gradient of geological conditions and the frequency of construction technology adjustments, the boundary is dynamically searched and delineated in the tunnel construction model. Based on each boundary, the tunnel construction model is divided into multiple thin-layer sub-models with spatial connection order.

3. The method for establishing a tunnel construction model according to claim 2, characterized in that, The step of dynamically searching and delineating boundaries in the tunnel construction model based on the gradient of geological condition changes and the frequency of construction technology adjustments includes: The sliding window algorithm is adopted to slide the window within the tunnel construction model interval corresponding to the recent construction section, and calculate the gradient of geological condition changes and the frequency of construction technology adjustment within the window. The recent construction section includes the excavated but unstable section and the section to be constructed. When the gradient of geological conditions or the frequency of construction process adjustments within the window exceeds the threshold, the boundary of the thin-layer sub-model is determined according to the corresponding window position. Adjust the position, length, and shape of the boundary according to the preset adjustment range.

4. The method for establishing a tunnel construction model according to claim 1, characterized in that, The steps of obtaining the current construction section information and determining the thin layer range of the model based on the current construction section information include: Based on the current location of the construction section and the tunneling speed, predict the range of construction progress within the preset time frame in the future; Based on the predicted construction progress range and combined with the trend of geological condition changes, the front and rear boundaries of the model thin layer range are dynamically determined. Based on construction monitoring data, the risk-affected areas are detected, and when adverse geological conditions or construction anomalies are encountered, the model's thin-layer range is automatically expanded to include the risk-affected areas.

5. The method for establishing a tunnel construction model according to claim 1, characterized in that, The step of extracting the corresponding current thin-layer sub-model based on the location interval of the thin-layer range in the tunnel construction model, and forming the current abbreviated model based on the model parameters of the current thin-layer sub-model includes: Based on the location interval of the thin layer range in the tunnel construction model, extract several consecutive thin layer sub-models as the current thin layer sub-model; Based on the surrounding rock grade, deformation rate, groundwater conditions, and construction stage in the current thin-layer sub-model, the current thin-layer sub-model is abbreviated to form the current abbreviated model.

6. The method for establishing a tunnel construction model according to claim 5, characterized in that, The step of establishing a real-time decision-driven module for making construction decisions based on the current abbreviated model includes: Based on the thin-layer sub-models contained in the current abbreviated model, establish the driving space of the real-time decision-driven module; Based on the spatial topology of each thin-layer sub-model in the current abbreviated model, a data transmission mechanism for each thin-layer sub-model is established in the real-time decision-driven module to realize the real-time sharing of construction monitoring data and predicted construction monitoring data of each thin-layer sub-model in the real-time decision-driven module; wherein, the spatial topology includes the forward connection relationship, backward connection relationship and lateral adjacent connection relationship of each thin-layer sub-model in the current abbreviated model; Establish a shared data storage area corresponding to the driving space of the real-time decision-driven module.

7. The method for establishing a tunnel construction model according to claim 6, characterized in that, The steps for anomaly risk identification and processing based on injected construction monitoring data, involving multi-source data fusion and data transmission in the real-time decision-driven module, include: Temporal variation data is extracted from construction monitoring data, including TBM sensor temporal data, surrounding rock deformation temporal data, support structure stress temporal data, environmental monitoring temporal data, geological sketch temporal data, and advanced geological prediction temporal data. The real-time decision-driven module performs multi-source feature fusion on time-series changing data, injects the fused multi-source features into the data storage area corresponding to the thin-layer sub-model where the working face is located, and transmits data to other thin-layer sub-models in the current abbreviated model according to the data transmission mechanism, generating predictive construction monitoring data and injecting it into the corresponding data storage area, thereby generating decision-driven data packages. Anomaly risk identification and processing based on decision-driven data packets.

8. The method for establishing a tunnel construction model according to claim 7, characterized in that, The step of transferring data to other thin-layer sub-models in the current abbreviated model according to the data transfer mechanism, and generating predictive construction monitoring data and injecting it into the corresponding data storage area, includes: Based on the forward connectivity, backward connectivity, lateral adjacency connectivity and corresponding connectivity weights of each thin-layer sub-model in the current abbreviated model in the spatial topology, as well as the fused multi-source features, the predicted construction monitoring data corresponding to each of the remaining thin-layer sub-models in the current abbreviated model are calculated to achieve data transmission. The predicted construction monitoring data corresponding to the other thin-layer sub-models in the current abbreviated model are fused into predicted multi-source features and stored in the data storage area corresponding to the other thin-layer sub-models in the current abbreviated model to achieve data injection.

9. The method for establishing a tunnel construction model according to claim 7, characterized in that, The steps for anomaly risk identification and processing based on decision-driven data packets include: The real-time decision-driven module inputs the decision-driven data package into the risk type prediction model and uses the risk type prediction model to predict the current construction status type. The construction status types include normal status, surrounding rock instability, water and mud inrush, collapse and roof fall, equipment failure and surface subsidence. Based on the predicted construction status type, identify whether there are any abnormal risks.

10. A tunnel construction model establishment system, characterized in that, The tunnel construction model establishment system applies the tunnel construction model establishment method according to any one of claims 1 to 9; the tunnel construction model establishment system includes: The abbreviated model module is used to acquire the established tunnel construction model, divide the tunnel construction model into multiple thin-layer sub-models with spatial connection order; acquire the current construction segment information, determine the thin-layer range of the model based on the current construction segment information; extract the corresponding current thin-layer sub-model based on the location interval of the thin-layer range of the model in the tunnel construction model, abbreviate the current abbreviated model based on the model parameters of the current thin-layer sub-model, and establish a real-time decision-driven module for making construction decisions based on the current abbreviated model; The identification module is used to establish an automated data interface for acquiring construction monitoring data, and to quickly inject the construction monitoring data generated during the construction process into the real-time decision-driven module. The construction monitoring data includes TBM sensor data, surrounding rock deformation monitoring data, support structure stress monitoring data, environmental monitoring data, geological sketch data, and advanced geological forecast data. Based on the injected construction monitoring data, the real-time decision-driven module performs multi-source data fusion and data transmission to identify and process abnormal risks. The decision-making module is used to obtain intelligent decision-making data based on the results of abnormal risk identification, and to push the intelligent decision-making data to each thin-layer sub-model in the current abbreviated model.

Images